measuring electronic latencies in minos with auxiliary detector

Measuring electronic latencies in MINOS with Auxiliary Detector

Son V. Cao University of Texas at Aus@n

(on behalf of the MINOS collabora@on) June 14th 2012

MINOS overview TOF principle AD setup Measurement Summary

FNAL2012 New Perspec@ves

2


Main Injector Neutrino Oscilla@on Search (MINOS)

734 km

Near det. 1.0 kton

Far det. 5.4 kton

ü High intensity muon neutrino beam produced at Fermilab ü  Two detectors: reduce systema@c errors, magne@zed to dis@nguish interac@ons

ü  Near Detector to measure ini@al beam composi@on, predict no-‐oscilla@on spectrum at Far Detector ü  Long baseline (734km) Time of

Flight (TOF) (1st measurement in 2007)

We revisit TOF with more staDsDcs and refined understanding of system

⌫µ and ⌫̄µ

3


v =d21

(t2 � t1)

Basic principle:

Real measurement:

P1(t1) P2(t2)

Distance d21

t1 t2

Find to minimize the likelihood between and

In 2007, used 473 FD events:

2. Time of Flight (TOF) principle

�t21P2(t2)P1(t1)

Need to understand the Dming system beNer!!!

Time shiP by TOFv

Baseline à TOFc

�t = �126± 32(stat.)± 64(syst.) ns (68% C.L. )

(v/c� 1) = 5.1± 2.9⇥ 10�5(68% C.L.)

Source of latency Uncertainty 2007

ND electronic latencies 32ns

FD electronic latencies 3ns

F/N detector rela@ve latencies 9ns

4

3. Auxiliary Detector (AD) MINOS overview TOF principle AD setup Measurement Summary

Auxiliary Detector at the back end of Near Detector

Auxiliary Detector in the middle of two Super-‐Modules at Far Detector

⌫ beam

Auxiliary Detector Dmestamps muon, resultant parDcle of neutrino interacDon, and matches with events recorded in real detector

ü  Readout system has complicated latency, not easy to measure

ü  Two readout systems of MINOS detector are not the same systema@c error

ü  Using two iden@cal readout systems, Auxiliary Detector (AD), can cancel out this error

5

Goal: measure the latencies and miDgate the systemaDc error using a simple auxiliary system

Near Far

UTC Dme

⌫ beam

AD AD

3. AD setup: mo@va@on MINOS overview TOF principle AD setup Measurement Summary

5. Event @mestamp with AD 3. AD setup: event @mestamp MINOS overview TOF principle AD setup Measurement Summary

6

ü  Passage of muon produces blue scin@lla@on light

ü  Wavelength-‐shi`ing fiber absorbs and shi`s scin@lla@on light

ü  Signals are mul@plied by photomul@plier (PMT) and used as “STOP” signal

ü  Timestamp “STOP” signal with a “START” reference signal in each detector

RelaDve electronic latency between ADs and real detectors is calculated by event matching

ScinDllator paddle

PMT

7

Timing diagram

Timing system overview MINOS overview TOF principle AD setup Measurement Summary

Beam structure ( 53MHz unit)

~10 us

ü  Accelerator generates signal, named $74, tell kicker magnet to prepare extrac@ng the proton beam

ü  Each spill lasts about 10us ü  At ND, Front-‐End boards start

window, labeled SGATE, to collect data

ü  At FD, another window, labeled

SS Window, is opened to capture data at the @me spill arrives

8

5. Matching AD and ND

Ending edge

Matching AD and ND time distribution (ns) 11100 11150 11200 11250

Num

ber o

f eve

nts/

2ns

0

500

1000

1500

2000

Auxiliary Detector

Near Detector

MINOS Preliminary

Dip Dip


Num

ber o

f eve

nts/

2ns

0

1000

2000Auxiliary Detector

Near Detector

MINOS Preliminary


Num

ber o

f eve

nts/

2ns

0

1000

2000

3000

Auxiliary Detector

Near Detector

MINOS Preliminary

Leading edge Ending edge Dip

Time relative to spill (18.83ns bins)200 400 600

Num

ber o

f eve

nts

/ 18.

83ns

0

100

200

300

400MINOS Preliminary


Matching between Auxiliary and Near Detector distribu@on

4. AD measurement at Near Beam structure ( 53MHz unit)

~10 us $74

SGATE

Time of μ at AD

Time of μ at ND

Use $74 and SGATE as “START” signals respec@vely for Auxiliary and Near detectors

8

AD data at Near

19ns

Bucket structure

9

Offset between SGATE and $74 at AD(ns)212850 212900 212950

Log

likeh

ood

0

2000

4000

6000

8000

MINOS PreliminaryDistribu@on comparison


4. AD measurement at Near

19ns

ü  Calculate likelihood of AD-‐ND distribu@on comparison as a func@on of SGATE-‐$74 “offset”

ü  Likelihood reflects very well the

bucket structure but the true peak is not clear.

ü  SGATE-‐$74 @me interval is not true

offset. SGATE is programmably lined up with 53.1MHz clock @ck.

ü  The direct measurement of SGATE-‐$74 @me interval also shows two peaks separated by 19ns.

SGATE -‐ $74 @me interval

Num

ber o

f events

10

Offset between SGATE and $74 at AD(ns)212850 212900 212950

Log

likeh

ood

0

2000

4000

6000

8000

MINOS Preliminary

Entries 5121

Mean 212908.120

RMS 16.877

Offset between SGATE and $74 at AD (ns)212850 212900 212950 213000

Num

ber o

f eve

nts

/ 2ns

0

50

100

150 Entries 5121

Mean 212908.120

RMS 16.877

MINOS Preliminary

Distribu@on comparison


4. AD measurement at Near

19ns

Event-‐by-‐event matching

No latency included

ND -‐ AD latencies = 36 +/-‐ 4ns

ü  AD is synchronized to ND DAQ, permined to match in event-‐by-‐event basis.

ü  These peaks agrees with those from

distribu@on comparison

ü  “Offset” measured by AD includes the electronic latencies of AD and ND

ü  Calculate AD-‐ND electronic latency by

comparing with @me interval from direct measurement (mean of the peak is used)

AD-‐ND latencies included

4ns

11

5. Matching AD and ND MINOS overview TOF principle AD setup Measurement Summary

4. AD measurement at Far

FD -‐ AD latencies = 12 +/-‐ 2ns

Entries 81Mean 98.602RMS 8.083

/ ndf 2r 57.218 / 47Constant 1.028± 23.342 Mean 0.102± 96.579 Sigma 0.123± 2.145

Offset between AD and FD (ns)60 80 100 120 140

Num

ber o

f eve

nts

/ 2ns

0

10

20

Entries 81Mean 98.602RMS 8.083


MINOS Preliminary

ü  Use cosmic ray events for matching

ü  Use PPS signals from Cs Clock and GPS as “START” signals respec@vely for Auxiliary and Far detectors

Cs PPS

GPS PPS

Time of μ at AD

Time of μ at FD

Monitored by Agilent 53230A

ü  The ADs are necessary for mi@ga@ng the electronic readout latencies

ü  The ADs has already installed and first results are used for TOF retrospec@ve analysis in 2012

ü  The AD will be used in the next phase of Time Of Flight measurement

12

Summary Summary

Source of latency Uncertainty 2007 Uncertainty 2012

ND electronic latencies 32ns 4ns

FD electronic latencies 3ns 2ns

F/N detector rela@ve latencies 9ns 1ns


Full Spill approach: TOFv – TOFc = -‐18 ± 11 (stat.) ± 29 (syst.) ns Wrapped Spill approach: TOFv – TOFc = -‐11 ± 11 (stat.) ± 29 (syst.) ns

13

5. 4. MINOS plans

13

Backup: Fit at peaks MINOS overview TOF principle AD setup Measurement Summary

Entries 5121Mean 212908.120RMS 16.877



Num

ber o

f eve

nts

/ 2ns

0

50

100

150Entries 5121Mean 212908.120RMS 16.877


MINOS Preliminary

Entries 5121Mean 212908.120RMS 16.877



Num

ber o

f eve

nts

/ 2ns

0

50

100

150Entries 5121Mean 212908.120RMS 16.877


MINOS Preliminary

14

5. 4. MINOS plans

2007 Analysis

Phase I (6 months)

Phase II (9-‐12 months)

Phase III (2013 and on MINOS+ era)

Re-‐analysis exis@ng data, reduce dominant systema@c errors

Hardware improvement

More hardware improvement , update beam energy

64 ns 18-‐33 ns 11-‐18 ns 2-‐7 ns

•  (A) Resurvey the distance between two detectors

•  (B,D) Re-‐es@mate the length of cables

•  (F) Bener synchroniza@on systems

•  (C,E,G) Auxiliary detector to measure delays of readout system

14

Backup: Systema@c revisit MINOS overview TOF principle AD setup Measurement Summary

15

5. Backup: two TOF approaches MINOS overview TOF principle AD setup Measurement Summary

•  NuMI neutrinos spill lasts about 10us •  Spill subdivided into 1.619 us batches (5 or 6 batches) •  95ns gap between batches

Full Spill approach: ü  ND event @me within spill

predicts FD @ming distribu@on ü  Find best value of Time of Flight

to match data

Wrapped Spill approach: ü  Wrap events within batch in each

detector ü  Fit two gaps in two detector and

subtract them

16

5. Backup: update @ming system

Use TWSTT* or PPP

*TWSTT: Two-‐Way Satellite Time Transfer PPS: Point Posi@on System

Near Far

Rb clock

•  Current GPS has 100ns jiNer •  Update precise and monitored sync

btw Near and Far det

MINOS overview TOF principle Recent TOF results MINOS plans AD setup Summary

From PMT to Amplifier to discriminator to Logic Unit

NIM 715: five-‐channel constant fracDon Dming

discriminator

NIM Model 612A: 12-‐Channel PhotomulDplier Amplifie

PMT 1

2 ns

2 ns

2 ns

2 ns

Model 364AL: Dual 4-‐Fold Majority Logic Units

1 ns

1 ns

1 ns

1 ns

8 ns

8 ns

8 ns

TDC TDC TDC

PMT 2

Model 1880B: Dual Visual Scaler

Design slightly different 17

measuring electronic latencies in minos with auxiliary detector

Education